Field of the Invention
The present invention relates to image capture apparatuses and methods for controlling the image capture apparatuses, and relates in particular to an automatic focus detection technique.
Description of the Related Art
Provision of a specific function to a part of a pixel group arranged in an image sensor in order to achieve automatic focus detection (AF) of a phase-difference detection type using a signal obtained from the image sensor has been known. For example, Japanese Patent Laid-Open No. 2000-292686 discloses an image sensor in which focus detection pixels provided with a pupil division function by arranging a light receiving region while shifting this light receiving region with respect to an optical axis of an on-chip microlens are arranged at prescribed intervals within a pixel region. With this image sensor, focus detection of the phase difference detection type can be performed based on a pair of signals obtained from focus detection pixels having different pupil division directions.
Meanwhile, recent image capture apparatuses use a method appropriate for application of signals when reading out the signals from pixels in an image sensor. For example, when performing live-view display, since the resolution of a display apparatus is lower than the number of pixels in an image sensor, signals of an appropriate number of pixels for the display are obtained by reading out the pixels while adding signals of a plurality of pixels, or reading out the pixels while thinning them out. When recording a moving image as well, similarly, an image of a moving image resolution is acquired. When shooting and recording a moving image as in the case of live-view display, smoothness of the display is more important than the resolution unlike at the time of recording a still image, and therefore it is desirable to increase the frame rate.
However, if the frame rate is increased, the exposure time per frame becomes shorter. For this reason, it is difficult to maintain the accuracy of automatic focus detection using signals obtained from the focus detection pixels, particularly for a low-luminance object. With regard to this problem, Japanese Patent Laid-Open No. 2008-085738 has proposed a method in which signals obtained from the focus detection pixels are added for a plurality of frames, and focus detection is performed using the added signal.
The relationship between a defocus amount and positional difference (phase difference) between a pair of signals obtained from the focus detection pixels varies in accordance with shooting conditions such as the f-number. Accordingly, a coefficient to be used when converting the positional difference into the defocus amount needs to be a value suitable for the shooting conditions. However, with the configuration described in Japanese Patent Laid-Open No. 2008-085738 in which the positional difference between a pair of image signals obtained by addition is converted into the defocus amount, the conversion into the defocus amount will be performed with one coefficient even in the case where the shooting conditions change during a time period of adding the signals of the focus detection pixels. Accordingly, it is difficult to obtain an accurate defocus amount.